Different cloud computing models exist based on the abstraction level of services, such as
newlineIaaS, PaaS, and SaaS that stand for Infrastructure as a Service, Platform as a Service, and
newlineSoftware as a Service, respectively facilitating pay-per-use basis for consumers. These
newlinemodels around services have been significant over a wide range of tasks, making cloud
newlinecomputing an extremely popular tool. This has enabled users to avoid upfront infrastructure
newlinecosts with over cloud computing option. Virtualization allows cloud providers to multiplex
newlineresource demands from numerous users. To remain profitable, cloud service providers must
newlineoperate at an elevated level of efficiency with large scale multiplexing. Cloud computing has
newlineresulted in huge data centers including millions of servers with an exponential rise in users.
newlineAt this magnitude, resource management has always been a significant issue. Scheduling
newlineoversees resource allocation and is crucial to resource management. The challenge of
newlineefficiency at this scale along with evolving models in cloud service consumption has
newlinenecessitated applications with methodologies and techniques aimed at replacing the age-old
newlinefewer effective issues in scheduling. In this research we present the challenges of scheduling
newlinevirtual computers specific to power and energy parameter. To make better scheduling
newlinedecisions, various fine-grained monitoring information is exploited over increasing
newlineeffectiveness of power and energy parameters.
newlineAs a scalable and adaptable solution to the widespread combination of devices like NoC
newlinequotNetwork-on-Chip , currently used in multi-core CPUs servers along with SoCs (System-on-
newlineChips). To deal with energy and reliability concern Self-aware multi-core architectures have
newlinebeen explored adaptive to changing demands from applications and users, with better
newlineperformance under energy constraints and resource availability. From the application (such
newlineas task mapping and scheduling) to the core (such as DVFS and power-gating) to the
newlineinterconnect fabric, this flexibility need can

Guide(s): KULKARNI, SUBHASH | Publisher: Belagavi | Handle: https://shodhganga.inflibnet.ac.in/handle/10603/715835

plus-circle Add Review

comment

Reviews